Positron emission tomography (PET), a noninvasive brain imaging technique, and psychophysiologic (e.g., EEG) studies are used to study biological correlates of drug abuse in human volunteers. PET and EEG studies show that abused euphoriants reduce global and regional cerebral metabolic rates for glucose (rCMRglc), and increase a power. Drug craving is an important factor in the relapse of patients with addictive disorders. In volunteers with histories of cocaine abuse, PET and EEG studies indicate that cocaine-related visual cues produce self-reports of cocaine craving, EEG arousal and increased rCMRglc in a widespread pattern throughout the cortex. Identification of such neural substrates of the responses to drug-related environmental stimuli may benefit development of therapy for cocaine dependence. Patients with acquired immunodeficiency syndrome (AIDS) show metabolic defects in the limbic system, consistent with an involvement of the hippocampus and related areas in cognitive deficits associated with AIDS dementia. Magnetic resonance imaging (MRI) is used to study brain structure, as related to personality and cognition in substance abusers. The orbitofrontal cortex is a brain region that shows a difference in rCMRglc in drug abusers as compared to controls. Volumetric MRI analyses on polysubstance abusers indicate that the inferior prefrontal lobe is smaller in abstinent abusers than in controls, but this difference apparently reflects a reversible change, as the asymmetry returns toward control with increasing length of abstinence. Results assessing the subjective and physiological effects of iv nicotine in smokers and nonsmokers suggest that nicotine increases heart rate and decreases rCMRglc in both groups whereas the subjective responses differ qualitatively. Smokers report positive feelings and liking the drug effect but nonsmokers cite disorientation, bad drug effects and a greater drug strength. A newly initiated protocol studying smokers will assess cognitive performance and regional cerebral blood flow during a cycle of nicotine deprivation. Development of a new model for calculating of (CMRglc) from PET data has been completed. This model eliminates the early phase of arterial blood sampling after radiotracer injection and simplifies data acquisition.